1. Technical Field
Present invention embodiments relate to data analysis, and more specifically, to automatic generation of an Extract, Transform, Load (ETL) job.
2. Discussion of the Related Art
Extract, Transform, Load (ETL) jobs extract data from a source, transform the extracted data using one or more transformations to a format compatible with a target, and load the data into the target, for example a target database. Existing tools to create ETL jobs are time consuming and complex, requiring specialized skills and training. Input data elements must be manually mapped to output data elements, and complex data transformations must be selected on a step-by-step basis.
To ease this process, existing tools often employ complex Graphical User Interfaces (GUI) requiring users to select transformations that convert input data into desired output data. However, users often do not know what transformations are needed, or how to set up a sequential series of transformations as may be required. Other existing tools require users to provide complex transformation metadata to the ETL job creation process, which is also complex and time consuming. For these reasons, businesses often hire outside experts to design and construct ETL jobs. Once the ETL job is created, future changes to maintain the job would again require specialized skills.